Taylor AJ, Villines TC, Stanek EJ, et al. Extended-release niacin or ezetimibe and carotid intima-media thickness. N Engl J Med. 2009. Epub ahead of print.
Randomized, unblinded intervention trial
317 patients with a history of coronary artery disease (CAD) or at high-risk for CAD who had been taking a statin drug (for 6+5 years) that had resulted in low-density lipoprotein cholesterol (LDL) levels <100 mg/dL and who had high-density lipoprotein cholesterol (HDL) levels <50 mg/dL for men or <55 mg/dL for women entered the trial, of whom 208 completed the trial.
Study Medication and Dosage
All patients continued to take statin drugs throughout the study. Patients were randomized to add either 10 gm/day Zetia (ezetimibe), a drug known to further lower LDL levels, or time-release niacin (TRN) with a target dose of 2,000 mg/day. Of those assigned to TRN, 75% achieved the 2,000 mg/day dosage, with the rest taking doses ranging from 500 mg/day to 1,500 mg/day. (While not specifically stated, it is likely that those taking lower doses did so as the result of intolerance to TRN at the full target dose.) The intervention lasted 14 months.
Primary Outcome Measures: Change in carotid intima-media thickness after 14 months, changes in lipid levels, and major cardiovascular events (e.g., myocardial infarction, revascularization, and death from CAD)
Mean HDL levels rose 7.5 mg/dL in the TRN group (P=0.001), while falling 2.8 mg/dL in the ezetimibe group. Mean LDL levels in the ezetimibe group declined by 17.6 mg/dL (P=0.001), compared with a 10.0 mg/dL decline in the TRN group (P=0.001). Significant reductions in triglyceride levels were also reported in both groups. Carotid intima-media thickness declined significantly in the TRN group (P=0.003) but not in the ezetimibe group.
The incidence of major cardiovascular events was 1% in the TRN group (2/160) versus 5% in the ezetimibe group (9/165) (P=0.04 for the difference). Of those events, 7 patients given ezetimibe died during the course of the trial compared with only 1 death in the TRN group. Three patients withdrew from the ezetimibe group due to drug side effects compared with 17 patients receiving TRN (NS).
Practice Implications: This is the first clinical trial to compare the effects of statins plus niacin (as TRN) with the effects of statins plus ezetimibe. The outcome was strikingly clear, with TRN coming out the winner.
Both arms of the trial continued to take statin drugs as they had been doing before the trial’s onset. Statin therapy alone has been shown to lower LDL and decrease the risk of coronary events. Historically, LDL levels are viewed as integral to the disease process and not simply as markers. Thus, most researchers and practitioners have believed that virtually anything that lowers LDL will lower the risk of CAD. Indeed the vast majority of agents or lifestyle changes that have been studied fit this pattern. These new findings therefore contradict our previously held beliefs; the researchers conclude, “ . . . our findings challenge the usefulness of LDL cholesterol reduction as a guaranteed surrogate of clinical efficacy.”
The conventional drug studied, ezetimibe, had previously been proven to add to the LDL-lowering effects of statins but had not been proven to achieve primary or secondary prevention of cardiovascular disease in the process. The new findings appear to suggest that ezetimibe-induced reductions in LDL buy no clinical benefit and may actually increase atherosclerosis (as indicated by the increase in carotid intima-media thickness) while quintupling the risk of cardiovascular events, at least when compared with the findings in the TRN group.
Nonetheless, these data provide no reason to consider ezetimibe a therapeutic addition to treatment with statin drugs.
The link between the extent of LDL-lowering and worse clinical outcomes in the ezetimibe group was uncovered in post hoc analysis. As a result, we should view the ezetimibe data with caution. Post hoc analyses often provide misinformation. If one hunts long enough, eventually by chance, statistical significance will attach to something. By chance alone, one observation in 20 should achieve statistical significance. Thus, the LDL-lowering effect of ezetimibe may not have been responsible for the apparent disease progression reported by these researchers in subjects for whom the decrease in LDL was most profound. Nonetheless, these data provide no reason to consider ezetimibe a therapeutic addition to treatment with statin drugs. Unlike ezetimibe, niacin (either in crystalline or time-release forms) is known to raise HDL levels while also reducing or stabilizing atherosclerosis and decreasing the risk of coronary events.
The amide form of vitamin B3 (niacinamide) is not used to treat patients with CAD because despite its excellent safety profile, it does not lower lipid levels. Crystalline niacin, which does lower LDL while raising HDL levels, is often avoided by cardiologists because of the high incidence of side effects both acute (e.g., gastrointestinal distress, headaches, flushing) and chronic (e.g., elevations in serum uric acid, glucose, liver enzymes). TRN avoids most (though not all) of the acute problems associated with standard (crystalline) niacin but has been linked with a higher incidence of hepatotoxicity.
While the difference in side effects between ezetimibe and TRN did not achieve statistical significance, it probably was the result of real problems caused by niacin and perhaps specifically by the time-release form of it. Patients assigned to large (500–2,000 mg/d) doses of TRN require monitoring for alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels as well as fasting glucose and uric acid levels.
These findings suggest that when possible (dependent upon the patient’s tolerance to TRN) TRN should now be added to statin therapy.